Alzheimer's disease (AD) is a neurodegenerative dementia characterized by selective cholinergic neurodegeneration. Millions of Americans are affected, but there are no disease-modifying treatments. This is, in part, because we do not know why the cholinergic neurons are more vulnerable than others, and therefore do not have specific strategy to protect them. In fact, there is even not an animal model that recaptures the selective and robust basal forebrain cholinergic (BFC) neuronal cell loss that is typical of AD. The absence of a model severely limits relevant mechanistic and therapeutic studies. We have now developed a transgenic mouse that exhibits the most essential features of cholinergic neurodegenerative process of human AD, including the robust BFC neuronal loss. These mice have G protein- coupled receptor kinase-5 (GRK5) deficiency (as severe as in human AD) and overexpress Swedish mutant of -amyloid precursor protein. The heterozygous double defective mice (hereafter abbreviated as GAP mice) selectively lost one-third of their BFC neurons at 18 months of age. We therefore propose that GRK5 deficiency is an Alzheimer-selective factor that makes cholinergic neurons more vulnerable to degeneration. GRK5 deficiency was previously linked to AD because it could be caused by -amyloid (A) and oxidative stress. There is a severe GRK5 deficiency in human AD brains. We now have preliminary findings that document that GRK5 deficiency increases cholinergic vulnerability both in cell cultures and in intact mice. GRK5 deficiency leads to reduced hippocampal acetylcholine release, cholinergic axonopathy (without cell death), and mild cognitive impairment. GRK5 deficiency also exaggerates Ass accumulation and gliosis. Mechanistically, all phenotypes of GRK5 deficiency appear to be attributable to an impaired desensitization of M2 muscarinic acetylcholine autoreceptor (M2). Extant research supports the concept that impaired M2 desensitization leads to persistent inhibition of the cAMP-dependent signaling pathway and that this inhibition reduces intrinsic defense mechanisms of cholinergic cells and leads to their vulnerability. We hypothesize that GRK5 deficiency selectively causes cholinergic vulnerability via impaired M2 desensitization; and that blocking presynaptic M2 receptors would prevent cholinergic neurodegeneration. We propose 4 Specific Aims to consolidate our preliminary findings and address our hypothesis. Specific Aim 1). Characterize the time course of cholinergic neurodegeneration and cognitive decline in GAP mice; Specific Aims 2 and 3). Compare the efficacies of M2 blockade and M1 stimulation in improving cognitive deficits and preventing cholinergic neurodegeneration in GAP mice; and Specific Aim 4). Translate the major findings from GAP mice to human AD by examining all these changes in human AD brain samples and correlating them with severity of dementia. We hope by the end of the project, we will have characterized an innovative animal model (GAP mice) of human AD for its detailed time course of cholinergic neurodegeneration and cognitive decline. We will have validated the innovative concept that GRK5 deficiency causes selective cholinergic vulnerability and that vulnerability can be prevented by M2 blockade but not M1 stimulation. In addition, at a principle level, we will have also proven the efficacy of a novel drug, AAD23, in preventing BFC neurodegeneration. We hope to find that the degree of GRK5 deficiency in human AD samples correlates with the severity of cholinergic neurodegeneration. This finding will provide rationale for trials of M2 receptor blockers for the prevention of cholinergic neurodegeneration in human AD.